Private LoRaWAN Network Gateways: Assessment and Monitoring in the Context of IIoT-Based Management ^†

Abstract

In the ever-evolving construction industry, the incorporation of the Industrial Internet of Things (IIoT) through Low-power Wireless Area Networks (LPWAN), such as LoRaWAN, has emerged as a practical solution for addressing the challenges posed by the limited 5G cellular coverage found in solutions like NB-IoT and LTE-M, especially when deployed in remote locations. Open-source LPWAN platforms like The Things Network (TTN) and ChirpStack have played a pivotal role in fostering the adoption of LoRa technology by providing a mature and cost-effective ecosystem that facilitates efficient device resource management. Within this context, maintaining continuous surveillance of LPWAN network gateways becomes critically important, requiring a meticulous examination of status indicators and an evaluation of the communication quality. This paper introduces a structured approach for extracting data from TTN to create a comprehensive gateway monitoring system. The methodology encompasses various aspects, including ensuring seamless server connectivity, specifically focusing on efficient information management and integration of real-world construction data. This foundational work sets the stage for a more in-depth exploration of the diverse management components within the network ecosystem.

1. Introduction

In the context of Industry 4.0, the construction industry is actively seeking innovative solutions to overcome operational challenges. Adopting the IIoT, particularly through private LoRaWANs, can revolutionize the industry by addressing communication, monitoring, and management issues [1]. However, effective gateway monitoring and management are essential to leverage private LoRaWANs’ benefits fully [2,3]. Monitoring LoRaWAN private gateways can bring advantages such as increased network reliability, proactive maintenance, data integrity, and optimal LoRaWAN network performance [4,5]. Furthermore, these methods can detect connectivity problems and enable corrective actions for seamless data flow, enhancing the infrastructure’s reliability, which is vital for IIoT devices. This work aims to provide a method for assessing and effectively monitoring gateways in private networks, specifically designed for an IIoT framework centered on industrial machinery management [6,7]. This endeavor represents a significant step toward addressing the pressing operational challenges faced by the construction industry in its quest for innovation and enhanced efficiency in the digital era of Industry 4.0.

Therefore, the contributions of this work are:

• The implementation of a method for active monitoring of gateways in private LoRaWAN networks based on the TTN platform.
• A data management agent for collecting, structuring, and sending data from the TTN server to third-party applications.
• Analysis of data extracted from monitoring the communication between private gateways and connected devices in a real industrial environment.

This manuscript is structured as follows: Section 2 reviews relevant literature; Section 3 discusses the proposed network monitoring architecture; Section 4 presents some evaluation results within the gateway scope; and Section 5 concludes and paves the way for future work.

2. Related Works

Table 1. Comparison of related works and approaches for monitoring and management of Private LoRaWAN Gateways.

Reference	Year	Contribution	Limitation and Challenges
[1]	2023	A new solution for real-time integrated monitoring of construction machinery in the construction industry.	The paper does not provide a detailed analysis of the solution’s performance in terms of operational efficiency, including, e.g., gateways.
[8]	2022	IIoT solution for a factory in the Manaus Industrial Complex that meets specific requirements such as MQTT message publishing gateway in a private broker, end-devices with adequate processing/storage capacity.	The system’s performance is evaluated based on a limited number of transmitted packets. Furthermore, the work needs to discuss the scalability of the system and its ability to handle many end devices.
[3]	2022	It provides original results comparing the performance of private and public LoRaWAN deployment options.	The paper does not provide an in-depth analysis of the deployment process for each approach.

depicts an analysis of a few recent related works, and, despite advances in the area, these studies have not implemented private gateway monitoring in the way this article addresses.

3. IIoT-Based Management Architecture

As depicted in Figure 1, the architecture establishes a construction machinery monitoring architecture rooted in the paradigm of IIoT. This structure revolves around end devices interfacing with machinery through industrial communication protocols. These devices employ LoRaWAN technology to transmit data extracted from machinery to dedicated management sites. Notably, the architecture relies on a private LoRaWAN network, emphasizing that gateways’ creation, configuration, customization, and administration require oversight from network administrators [8,9].

The Things Network, in its community version, provides a long-range, low-power communication framework finely tuned to the needs of Internet of Things (IoT) devices and applications that utilize LoRaWAN. This platform embodies an open and collaborative infrastructure, empowering users to establish personalized IoT networks in the various locations that the construction industry demands. Using LoRaWAN private networks in TTN allows the management of their gateways through procedures for the creation, configuration, deletion, monitoring, validation, and user authentication, among others.

3.1. TTN Server and Integrators

The function of integrating devices with the consumer’s application is carried out by TTN integrators, serving as intermediaries, generating connections between devices’ events/messages and third-party applications. This role assumed by integrators facilitates the seamless transmission of data acquired by IoT devices to designated management or visualization platforms. The integration can be performed by utilizing Application Programming Interfaces (APIs) and MQTT connectors provided by TTN as part of their services. The choice of the integrator towards TTN can influence the latency, reliability, adaptability, transformation, communication protocol, and connection of data to the platforms; therefore, the type of integrator must be related to the monitoring needs of the different events and messages available for each device or gateway.

3.1.1. MQTT Application and Gateway Server

In the TTN architecture, the MQTT Application Server and MQTT Gateway Server play distinct yet crucial roles. The MQTT Application Server connects IoT devices and applications, routing converted device data to specific applications and managing data subscriptions and publications from external applications. On the other hand, the MQTT Gateway Server interfaces with network-connected gateways, receiving, processing, and distributing gateway-transmitted data to TTN, ensuring smooth data flow in TTN’s IoT ecosystem. Furthermore, gateways occasionally send status updates to the MQTT Gateway Server after long periods; in the context of monitoring, it may be less effective to just rely on this source of status information.

Therefore, to monitor the private gateways, TTN’s MQTT Application Server grants access to crucial IoT device uplink data, like RSSI and SNR, reflecting connection quality. Meanwhile, the MQTT Gateway Server provides gateway performance data, availability, and connectivity status. It also enables gateway management and coordination for infrastructure control [10,11].

3.1.2. TTN Application and Gateway APIs

TTN’s application and gateway APIs are available via HTTP GET and POST requests. GET requests to specific URLs offer insights on device events and allow tailored device commands. Similarly, Gateway API GET requests provide gateway details, including status and performance metrics. Conversely, POST requests enable command distribution and gateway data retrieval. For gateway health monitoring, GET requests to relevant URLs yield real-time data on availability, including packet timestamps, connection status, version, model, and more. These data support alert systems and visualizations, ensuring prompt issue detection and optimal gateway operation in TTN [10].

3.2. Data Management Agent

The data management agent is pivotal in facilitating structured connectivity and oversight of acquired parameters via MQTT or API queries. This agent encompasses a publication and subscription service for gathering data from end devices. For gateway monitoring, it archives device-specific information accompanying uplink messages, including sender identity, receiving gateway identifier, and metrics such as RSSI, SNR, bandwidth, spreading factor, data rate index, and frequency. Moreover, the agent encompasses a service to query TTN APIs, offering insights into status, protocol, connection history, update frequency, message count since the last update, device model, and utilized frequencies. These data agent programmed with Python allows the management of TTN in a customized and non-standardized way. Therefore, it allows us to adapt to different sources and information from TTN, perform more complex processes, and manage the multiple destinations requiring the TTN’s components data. This agent’s role encompasses rectifying erroneous, atypical, or missing data, precluding disruptive application queries, and ensuring gateway monitoring stability. It optimizes database storage by retaining pertinent gateway monitoring data, excluding redundant information in the MQTT server and API queries. Crucially, this agent safeguards against the gateway status information gap by autonomously soliciting the current gateway status during device communication without relying solely on incoming messages.

4. Gateways Monitoring Assessment

This section shows the results obtained through the monitoring of private gateways. These gateways have been installed to collect data from a construction scenario in Vilar Formoso, Portugal, so the data represents the operation in a real IIoT scenario. Two private gateways were installed, a LORIX One LoRaWAN base station at the Vilar Formoso construction site and a LtAP LR8 LTE Mikrotik kit (MikroTik, Riga, Latvia) in a semi-fixed station in a crusher 12 km from the initial station at Porto de Ovelha, Portugal. These were intended to have the best coverage in a more than 30 km construction line. Therefore, due to the locations of the private gateways and their difficult access, creating an effective monitoring system is mandatory. First, an alert system was created to receive an email notification when a gateway was disconnected. In addition, this system displayed an updated status of the gateways to the construction operators and network management, allowing them to verify whether the gateways were operational or not.

In an actual industrial setting, the dataset includes transmission data from machines like a Multifunction, a Jaw Crusher, a bulldozer Dossan DL200, and a Drill Roc D7. The system uses the data management agent described in previous sections, which systematically collects and organizes device-specific data and gateway information, and the datasets comprise around two months of tests. As part of the monitoring tasks, the distribution of messages at each gateway is required; it is important to evaluate if the allotment and location of the gateways around the construction site are correct. Figure 2 shows message distribution among gateways and each machine, enabling analysis of gateway usage for each machine in our management platforms. In general, the use of the gateway far from the main site (Crusher Station) is greater for most machines, which may be due to its location and where the machines work. Another vital aspect of gateway monitoring is assessing communication quality between the machines and gateways. This dataset allows in-depth analysis for identifying issues related to LoRaWAN network coverage that affect the construction zones communication and eventually the loss of system reliability. Figure 3 demonstrates how gateway monitoring, combined with device in-formation, enables the evaluation of the average RSSI for each gateway–machine pair, aiding in the detection of abnormal data points. This insight helps in defining strategies to prevent future device-to-gateway communication issues. Finally, Figure 4 shows the variation in the communication quality between the gateways and machines over the distance from the antenna. It can be observed that the central station tolerates longer-distance communications, with a noticeable degradation when communication reaches 15 km. In contrast, the Crusher station does not tolerate longer communication distances, presenting a degradation and a maximum tolerance below 9 km of distance. Despite receiving a higher quantity of messages, its tolerance for longer distances can be diminished by its location in a working area, which typically has a less accessible topography.

5. Conclusions and Future Works

Integrating IIoT technology, such as implementing networks like TTN, holds trans-formative potential for revolutionizing equipment management and utilization strategies within the construction domain. Providing real-time connectivity, monitoring capabilities, and comprehensive data collection engenders more informed decision-making processes, optimization of resources, and operational workflows. The continuous monitoring of gateways emerges as a pivotal imperative to ensure connectivity and check performance standards within the TTN network. Effective management of gateway status and communication quality remains essential to uphold network stability and reliability. This study demonstrates an efficacious approach to resource monitoring, particularly on gateways, by harnessing data acquisition via MQTT or relevant requests directed at the TTN server. Moreover, the work establishes prerequisites for data cleaning, organization, and compaction, optimizing data transmission to visualization platforms and third-party entities. This system belongs to a complete scheme for IIoT-integrated management for the construction industry; hence, this work establishes a foundation for systematic and orderly data collection, laying the groundwork for future in-depth analyses. This study mainly shows the data obtained from monitoring the communication performance, mainly from uplink messages, as well as the status of the gateways. However, it does not consider monitoring the active management of resources and gateways, such as creation, update, deletion, purge, and authentication. Future work will consider the implementation of some of these features. Also, future implications can encompass a comprehensive assessment of the integrated device, machinery, operator, and network management; it provides a structured approach for systematically acquiring data to facilitate more profound investigations in subsequent research and works.